The potential of cardiovascular risk factors for reducing visual impairment: a pooled analysis of European epidemiological studies

Abstract

Purpose : To estimate the proportion of visual impairment (VI) that could potentially be avoided if the population was not exposed to cardiovascular risk factors.

Methods : The European Eye Epidemiology (E³) consortium is a collaborative network of epidemiological studies. Fourteen cross-sectional population-based studies from 9 European countries were included in a pooled analysis of individual participant data. VI was defined as best-corrected visual acuity < 20/60 at better eye. Cardiovascular risk factors included smoking, diabetes, hypertension, obesity and overweight, hyperlipidemia and history of cardiovascular disease. Multivariate mixed logistic regression models, using a random effect for studies, were used to estimate the odds-ratios of the associations of cardiovascular risk factors with VI. The proportions (95% CI) of VI due to the risk factors were estimated using population attributable fractions (PAF). PAF were calculated for statistically significant risk factors from the odds-ratios and the prevalence of the risk factors in subjects with VI, and their confidence intervals (CI) were estimated by bootstrapping.

Results : 55,467 subjects, aged 45 years or more, were included in the analysis. In the multivariate analysis including all risk factors, higher risk of VI was significantly associated with age (p<0.0001), female gender (p=0.02), smoking (p=0.0002), diabetes (p=0.001) and cardiovascular disease (p<0.0001), while decreased risk of VI was significantly associated with secondary and higher education (p=0.0002), and overweight and obesity (p=0.01). No statistically significant associations were found with hypertension and hyperlipidemia. PAF was 4.8% (95% CI: 2.7-6.8) for current smoking, 4.6% (95% CI: 2.9-6.4) for diabetes, 6.2% (95% CI: 3.5-8.8) for cardiovascular disease and 12.9% (95% CI: 9.5-16.5) for any of these 3 risk factors. PAF for cardiovascular risk factors were higher for men (23.7% for any risk factor, 95% CI: 16.7-31.0) than for women (9.3%, 95% CI: 5.2-12.7).

Conclusions : Our study shows that VI might be reduced by more than 10% if cardiovascular risk factors could be avoided, with a greater effect in men (23.7%) than in women (9.3%).

This is an abstract that was submitted for the 2017 ARVO Annual Meeting, held in Baltimore, MD, May 7-11, 2017.